Summary The majority of pancreatic cancers are driven by oncogenic KRAS. Even with this knowledge, there is an urgent need to identify vulnerabilities of this cancer, as the prognosis remains dire. We discovered a previously unrecognized function of STAT3 in the maintenance of epithelial cell identity and differentiation in cancer. In an experimental model, we found the loss of STAT3 in KRAS-driven tumors produced a mesenchymal-like identity, a phenotype known to be associated with metastasis, and drug-resistance. In the context of oncogenic KRAS, STAT3 is responsible for shaping tumor identity and epithelial differentiation. Mutant KRAS is a potent oncogenic driver, but recent evidence indicates that KRAS may not be necessarily required for tumor maintenance. To study the consequences of KRAS-independent tumorigenesis, we extinguished the expression of oncogenic KRAS in pancreatic cancer cells and found cell populations were still be able to form tumors. Most important, if STAT3 activity is lost in these KRAS-independent tumor cells, tumorigenesis is impaired. This is a novel finding and has significant implications for the treatment of pancreatic cancers. The overarching objective of this proposal is to elucidate the actions of STAT3 in cancers that are dependent on oncogenic KRAS, and cancers that adapt to KRAS independence. STAT3 is a transcription factor, and it is our hypothesis that STAT3 regulates genes that control differentiation, tumorigenicity, and immunity to influence cancer development and maintenance. The first specific aim will determine the mechanisms by which STAT3 regulates tumor identity and differentiation in KRAS pancreatic cancers. STAT3-driven transcription, pathway activation, and cell fate transitions will be evaluated. The second specific aim will define mechanisms by which STAT3 promotes the tumorigenicity of pancreatic tumors that have lost dependence on KRAS. Key genes will be defined that drive cancer stem cell maintenance, chromatin accessibility, and immune landscape. The broad reach of STAT3 is revealed in KRAS-independent tumors. The accomplishment of our aims is expected to reveal the impact of STAT3 on KRAS-dependent and KRAS-independent pancreatic cancer, and to provide knowledge that will lead to the development of more effective therapies.